Matrix Games Forums

Nelsonxu3320 wrote: Tue Sep 27, 2022 6:44 pm Hi,

I'm just so curious about the Combat/Mission Control system modelled in CMO after hundreds hours of gameplay. What it takes to build a system like this (Both in software and hardware)? What elements are omitted on purpose to simplify the overall complexity compared to the field-deployable system? What are the untold mechanics among the OODA cycle? I mean it can't be that simple by only 1-click to give a command to any of the fire units to conduct a precision strike on the designated target which is spotted by any reconnaissance platforms, right?

wirthlin wrote: Wed Sep 28, 2022 8:34 pm

Nelsonxu3320 wrote: Tue Sep 27, 2022 6:44 pm Hi,

I'm just so curious about the Combat/Mission Control system modelled in CMO after hundreds hours of gameplay. What it takes to build a system like this (Both in software and hardware)? What elements are omitted on purpose to simplify the overall complexity compared to the field-deployable system? What are the untold mechanics among the OODA cycle? I mean it can't be that simple by only 1-click to give a command to any of the fire units to conduct a precision strike on the designated target which is spotted by any reconnaissance platforms, right?

Hi Nelsonxu3320. You're asking good questions. Welcome to the wonderful world of modeling and simulation using C:MO. For an understanding of what it takes to build a system like this, I'm going to recommend getting a basic book on modeling and simulation. I've been looking at "What Every Engineer Should Know About Modeling and Simulation" by Madachy and Houston and published by CRC Press. It's meant for folks who have little or no experience in modeling and simulation and would seem to be a good starting point. Perhaps someone could recommend a better starting text.

While it won't give you specifics on the underlying C:MO architecture and design, it will give you an idea of the basic concepts that seem to be at work in C:MO.

I hope this helps a bit.

Nelsonxu3320 wrote: Wed Sep 28, 2022 9:27 pm

wirthlin wrote: Wed Sep 28, 2022 8:34 pm

Nelsonxu3320 wrote: Tue Sep 27, 2022 6:44 pm Hi,

I'm just so curious about the Combat/Mission Control system modelled in CMO after hundreds hours of gameplay. What it takes to build a system like this (Both in software and hardware)? What elements are omitted on purpose to simplify the overall complexity compared to the field-deployable system? What are the untold mechanics among the OODA cycle? I mean it can't be that simple by only 1-click to give a command to any of the fire units to conduct a precision strike on the designated target which is spotted by any reconnaissance platforms, right?

Hi Nelsonxu3320. You're asking good questions. Welcome to the wonderful world of modeling and simulation using C:MO. For an understanding of what it takes to build a system like this, I'm going to recommend getting a basic book on modeling and simulation. I've been looking at "What Every Engineer Should Know About Modeling and Simulation" by Madachy and Houston and published by CRC Press. It's meant for folks who have little or no experience in modeling and simulation and would seem to be a good starting point. Perhaps someone could recommend a better starting text.

While it won't give you specifics on the underlying C:MO architecture and design, it will give you an idea of the basic concepts that seem to be at work in C:MO.

I hope this helps a bit.

Hi wirthlin. Thanks for the tips. Actually, I am currently using MATLAB Simulink to model such a system that could be used on some MRLS or TBMS. The current step I am working on is to build a state machine modelling the interrelation between sequential behaviors of the operators and the information flow generated by both parent level and child level.
The limitations are obvious. The networking or data transmission can't be modelled as accurate as the real. So, I better study more on the networking part (like data links, data transmissions...) In my opinion, many of the fire control/mission control systems are just tons of calculators plus tons of networking links. To build a network of links is not an easy task. However, without the links, the calculators are meaningless. C:MO gives a hint, the hint can't show the whole picture, but still.

wirthlin wrote: Thu Sep 29, 2022 4:28 am

Nelsonxu3320 wrote: Wed Sep 28, 2022 9:27 pm

wirthlin wrote: Wed Sep 28, 2022 8:34 pm

Hi Nelsonxu3320. You're asking good questions. Welcome to the wonderful world of modeling and simulation using C:MO. For an understanding of what it takes to build a system like this, I'm going to recommend getting a basic book on modeling and simulation. I've been looking at "What Every Engineer Should Know About Modeling and Simulation" by Madachy and Houston and published by CRC Press. It's meant for folks who have little or no experience in modeling and simulation and would seem to be a good starting point. Perhaps someone could recommend a better starting text.

While it won't give you specifics on the underlying C:MO architecture and design, it will give you an idea of the basic concepts that seem to be at work in C:MO.

I hope this helps a bit.

Hi wirthlin. Thanks for the tips. Actually, I am currently using MATLAB Simulink to model such a system that could be used on some MRLS or TBMS. The current step I am working on is to build a state machine modelling the interrelation between sequential behaviors of the operators and the information flow generated by both parent level and child level.
The limitations are obvious. The networking or data transmission can't be modelled as accurate as the real. So, I better study more on the networking part (like data links, data transmissions...) In my opinion, many of the fire control/mission control systems are just tons of calculators plus tons of networking links. To build a network of links is not an easy task. However, without the links, the calculators are meaningless. C:MO gives a hint, the hint can't show the whole picture, but still.

Thank you for providing the context! You are looking for something much deeper than I assumed, my apologies.

As you pointed out, the data transmission or data link is difficult to model accurately. Especially if you don't have all of the needed information. In addition to modeling the data exchange protocols between transmitter and receiver, there is also the need to model the RF environment to determine whether or not there is even the possibility of communications. For instance, for frequencies at VHF and above, the receiver better have line of sight to the transmitter or at least to a relay such as a tower or a satellite. For HF frequencies, depending on frequency, time of day, and other factors, communications are possible worldwide without the need for LOS.

But how accurate does the data transmission model need to be for your purposes? For instance, the operator may not care about the details of how he get's the contact data. just that he gets it and then responds to it accordingly based on standing orders. The contact data may be complete, partial, false, or non-existent (no comms). In this case, is there a need to model the network in great detail?

I hope this helps out a little bit.

Nelsonxu3320 wrote: Thu Sep 29, 2022 6:53 pm

wirthlin wrote: Thu Sep 29, 2022 4:28 am

Nelsonxu3320 wrote: Wed Sep 28, 2022 9:27 pm

Hi wirthlin. Thanks for the tips. Actually, I am currently using MATLAB Simulink to model such a system that could be used on some MRLS or TBMS. The current step I am working on is to build a state machine modelling the interrelation between sequential behaviors of the operators and the information flow generated by both parent level and child level.
The limitations are obvious. The networking or data transmission can't be modelled as accurate as the real. So, I better study more on the networking part (like data links, data transmissions...) In my opinion, many of the fire control/mission control systems are just tons of calculators plus tons of networking links. To build a network of links is not an easy task. However, without the links, the calculators are meaningless. C:MO gives a hint, the hint can't show the whole picture, but still.

Thank you for providing the context! You are looking for something much deeper than I assumed, my apologies.

As you pointed out, the data transmission or data link is difficult to model accurately. Especially if you don't have all of the needed information. In addition to modeling the data exchange protocols between transmitter and receiver, there is also the need to model the RF environment to determine whether or not there is even the possibility of communications. For instance, for frequencies at VHF and above, the receiver better have line of sight to the transmitter or at least to a relay such as a tower or a satellite. For HF frequencies, depending on frequency, time of day, and other factors, communications are possible worldwide without the need for LOS.

But how accurate does the data transmission model need to be for your purposes? For instance, the operator may not care about the details of how he get's the contact data. just that he gets it and then responds to it accordingly based on standing orders. The contact data may be complete, partial, false, or non-existent (no comms). In this case, is there a need to model the network in great detail?

I hope this helps out a little bit.

No apologies needed here. All your points are solid and worth to keep in mind.

Well, for data link modelling, it really depends on the types of data carried by the links. If there are many cross types of data needed to be transmitted, no doubt for both the transmitter and receiver to have tasks heavier than others with more singular data types. In addition, the RF environment as you stated is also a factor. These factors will determine the accuracy (If the data can be transmitted and received faithfully) and time sensitivity (If the time duration needed sufficient for the mission) of the links. For example, there is a minor loss even with cabling. And there is a military standard (probably MIL-STD-1553B) regulates this.

For the example of the operator you gave, you talk about the pre-existed variances or errors. This is a brilliant point but a bit overextending for now. Nonetheless, this leads to another topic to think about: To what extent, the combat/mission system needs to process the raw data received before sending them? If the mission control just sends all the data received from all the sensors to the firing units, the operators will get overwhelmed even if all the info are 100% real, and it is more likely to cause some human errors. This is to say, there is a balance the parent level system and child level system to process the amount of data received. If the parent level compute too much, it will generate a fire solution and send it to the firing units. The only thing for the units to do it follow the solution and fire. But if the parent only allocates the target, the units is responsible for calculation of the solution. So, this is versus of centralized and de-centralized. However, to check if the data is complete, partial, false or whatever is far more complicated. This probably needs some machine learning algorithms and sensor fusion techniques to interpret and determine. That's why I think this is a bit beyond for now.

Anyway, I will see how it goes. There will be some field tests after the preliminary design. let you know if there are updates, and you are still interested.